Advanced Memory (Memops)

• Why — Give an MCP-capable assistant a durable place for notes, research, and retrieval: search and RAG over your content instead of losing context every session.
• What — A FastMCP 3.2 GA server (Python 3.12+) exposing 79 tools across 12 Managed Namespaces (audio, inbox, skills, zettel, nav, notes, search, knowledge, project, system, mcp, typora) covering web/GitHub/arXiv research, document ingestion, vector search (LanceDB), exports (HTML, PDF, Pandoc), skills workflows, and hooks into common note stacks (Obsidian, Joplin, Notion-oriented flows).
• How — Connect the MCP server from your client (installation, then usage). Optionally run the webapp for a browser UI on top of the same backend.

Documentation

• Installation — Python uv, optional OCR/Pandoc
• Usage — MCP clients, webapp, advanced topics
• AI features — RAG, agentic mode, sampling
• FastMCP 3.2 — Managed Namespaces, prefabs, CodeMode, transports
• Product requirements (PRD) — current mission, scope, KPIs (1.8.x)
• Architecture
• Fleet / multi-node
• Compliance & standards
• Development — just recipes (lint, test, pack)
• Changelog
• Release checklist (MCPB + Git tag)

Author: Sandra Schipal · Vienna, Austria

🛡️ Industrial Quality Stack

This project adheres to SOTA 14.1 industrial standards for high-fidelity agentic orchestration:

• Python (Core): Ruff for linting and formatting. Zero-tolerance for print statements in core handlers (T201).
• Webapp (UI): Biome for sub-millisecond linting. Strict noConsoleLog enforcement.
• Protocol Compliance: Hardened stdout/stderr isolation to ensure crash-resistant JSON-RPC communication.
• Automation: Justfile recipes for all fleet operations (just lint, just fix, just dev).
• Security: Automated audits via bandit and safety.